AI-Assisted Feedback for Learning Standardized Tuina Skills

Summary

The goal of this educational study is to determine whether personalized feedback generated using sensor data and generative artificial intelligence (AI) can improve the learning of standardized Tuina skills among rehabilitation trainees. Tuina is a form of manual therapy that requires learners to control the location, force, rhythm, and consistency of their hand movements.

A total of 81 rehabilitation trainees will be randomly assigned to one of three training groups: AI-supported personalized feedback, sensor-based data feedback without AI-generated recommendations, or traditional instructor feedback. All groups will receive the same standardized demonstration, training tasks, practice duration, and number of practice sessions.

The main question is whether trainees receiving AI-supported personalized feedback achieve better retention of standardized Tuina skills four weeks after training. The researchers will also compare immediate skill performance, force and rhythm control, transfer of skills to a related task, learning efficiency, self-efficacy, cognitive load, satisfaction, and the safety and acceptability of AI-generated feedback.

Skill performance will be assessed using a blinded Objective Structured Clinical Examination (OSCE) and objective sensor-based measurements. The study activities will be conducted using a mechanical simulation model and a pressure-sensing system, rather than on patients.

Conditions

• Medical Education
• Clinical Skills

Interventions

BEHAVIORAL

AI-Supported Personalized Feedback

Participants will complete four standardized training sessions over two weeks, with each session lasting approximately 45 to 60 minutes. A pressure-sensing system will measure force accuracy, force variability, operating frequency, rhythm stability, and time within the target range. After each practice attempt, a generative AI system will provide feedback using a structured task-gap-action format. The feedback will describe the target task, identify differences between measured performance and the predefined standard, and recommend specific actions for the next practice attempt. AI output will be restricted to educational use and overseen by instructors.

BEHAVIORAL

Sensor-Based Data Feedback

Participants will complete four standardized training sessions over two weeks, with each session lasting approximately 45 to 60 minutes. After each practice attempt, participants will view numerical sensor measurements, force-time curves, rhythm information, and predefined target ranges. No generative AI interpretation, personalized action plan, or AI-generated recommendation will be provided.

BEHAVIORAL

Traditional Instructor Feedback

Participants will complete four standardized training sessions over two weeks, with each session lasting approximately 45 to 60 minutes. Instructors will observe performance and provide conventional verbal feedback based on the standardized teaching protocol. Participants will not receive sensor-derived performance displays or AI-generated feedback.

Sponsors & Collaborators

• Zhejiang University

  lead OTHER

Principal Investigators

• Xing-Chen Zhou · The First Affiliated Hospital, Zhejiang University School of Medicine, 79 Qingchun Rd., Shangcheng District, Hangzhou, Zhejiang

Study Design

Allocation

RANDOMIZED

Purpose

OTHER

Masking

SINGLE

Model

PARALLEL

Eligibility

Min Age

18 Years

Sex

ALL

Healthy Volunteers

Yes

Timeline & Regulatory

Start

2026-08-13

Primary Completion

2027-05-01

Completion

2028-08-01